Revisiting the Mississippi Interracial Marriage Poll
I’ve taken a lot of heat in the comments for arguing that the recent poll showing a plurality of Mississippi voters would support a ban on interracial marriage is, in effect, bogus. I think a lot of that heat is deserved in retrospect. I’m still skeptical of the poll, so this isn’t a full retraction: the practice of throwing a single unrelated question, without any forewarning, at the end of a lengthy series of questions about political candidates still strikes me as highly unreliable and likely to create confusion in a respondent. However, I can’t prove that this is the case, and several commenters pointed out external evidence that would tend to support the poll’s results.
Specifically, one commenter noted that within the last decade, 40% of voters in neighboring Alabama actively voted against formally amending the state constitution to permit interracial marriage. Others pointed out that the high percentage of Mississippi Republicans who are over the age of 65 meant that about a third of Mississippi Republicans grew up at a time when interracial marriage was still prohibited.
Again, I’m still a bit skeptical of this poll’s results, and at a minimum I’d like to see some research into whether inserting an entirely random polling question at the end of a survey like that will affect how people respond, especially when you are dealing with an especially tedious robo-call. But the actual results of the poll are sufficiently consonant with circumstantial evidence that it probably shouldn’t have set my BS detector off so much in the first place.